ANTIOXIDANTS & REDOX SIGNALING
Volume 9, Number 9, 2007
© Mary Ann Liebert, Inc.
DOI: 10.1089/ars.2007.1660
Forum Original Research Communication
Contribution of Lysosomal Vesicles to the Formation of
Lipid Raft Redox Signaling Platforms in Endothelial Cells
SI JIN, FAN YI, and PIN-LAN LI
ABSTRACT
We have demonstrated that the formation of lipid raft (LR)-redox signaling platforms membrane is associated with activation of acid sphingomyelinase (ASMase) in coronary arterial endothelial cells (CAECs). Given
that the trafficking of lysosomal vesicles might play an essential role in ASMase activation, the present study
tested whether lysosomal vesicles contribute to the formation of LR redox signaling platforms. By confocal
microscopy, we found that Fas ligand (FasL) induced the formation of LR clusters in the plasma membrane
of CAECs, accompanied by aggregation of NAD(P)H oxidase subunits, gp91phox and p47phox, and ROS production. When the cells were pretreated with two structurally different lysosomal vesicle function inhibitors,
bafilomycin A1 (Baf) and glycyl-L-phenylalanine--naphthylamide (GPN), the FasL-induced LRs clustering
was substantially blocked, and corresponding ROS production significantly decreased. By confocal microscopic observations in living CAECs by using LysoTracker, a colocalization of LRs and lysosomal vesicles was
found around the cell membrane, which was abolished by Baf or GPN. Functionally, FasL-induced inhibition
of endothelium-dependent vasorelaxation was also reduced by both inhibitors of lysosome function. These results suggest that lysosomal vesicles importantly contribute to the formation of LR-redox signaling platforms
and thereby participate in the oxidative injury of endothelial function during activation of death receptor-Fas
in coronary arteries. Antioxid. Redox Signal. 9, 1417–1426.
L
(LRs) consist of dynamic assemblies of cholesterol and sphingolipids such as sphingomyelin (SM),
ceramide, glycosphingolipids, and gangliosides as well as some
phospholipids (2). Recently, increasing evidence exists that LRs
clustering as a general mechanism plays an essential role in the
initiation of receptor-mediated transmembrane cell signal transduction in a variety of mammalian cells such as lymphocytes,
epithelial cells, neurons, and endothelial cells (9, 14, 16, 28).
This LRs clustering-associated signaling mechanism is now
considered a novel model that transduces and amplifies the signals generated by activation of different receptors such as death
receptors, insulin receptors, some neurotransmitter receptors,
and growth-factor receptors (2). By facilitating or amplifying
transmembrane signaling, LRs clustering in the cell membrane
could regulate various cell or organ functions or cellular activIPID RAFTS
1Department
ities such as cell proliferation, differentiation and apoptosis, Tcell activation, tumor metastasis, and neutrophil or monocyte
infiltration and infection, and so on (2).
In recent studies, we reported that LRs clustering on the arterial endothelium recruited or aggregated some redox signaling molecules such as NAD(P)H oxidase subunits and Rac
GTPase. The aggregation of these redox signaling molecules
activated and enhanced production of superoxide (O2) or reactive oxygen species (ROS) and thereby impaired endothelial
function. This LRs clustering with redox signaling molecules
was referred as to the formation of LR redox signaling platforms (32). It has been demonstrated that these signaling platforms in the membranes of coronary arterial endothelial cells
(CAECs) are characterized by gp91phox aggregation, p47phox recruitment or translocation, and by activation of acid sphin-
of Pharmacology & Toxicology, Medical College of Virginia Campus, Virginia Commonwealth University, Richmond, Virginia.
1417
1418
JIN ET AL.
gomyelinase (ASMase) and subsequent production of ceramide.
The production of ceramide in the cell membrane may form
ceramide-enriched microdomains, which are able to fuse spontaneously to one or a few large macrodomains: clusters or platforms (15). However, it remains unclear how ASMases are recruited and activated in these LR platforms.
In previous studies, the rapid activation and translocation
of ASMases into LRs were observed in response to various
stimuli such as FasL, TNF- and endostatin (2, 32). However,
it remains unknown where these ASMase molecules come
from. Although it has been reported that the ASMase gene
gives rise to a common mannosylated precursor protein,
SMase precursor-mannose, this gene product is shuttled into
either the lysosomal trafficking pathway mediated by sortilin
and mannose 6-phosphate receptor (19) or a secretory pathway (27). Given evidence that the activated form of ASMase
localizes at the outer leaflet of the cell membrane (12), it
seems that lysosome-associated vesicle transportation or trafficking of ASMase from the abluminal compartment to the
cell membrane is a prerequisite for ASMase activation in the
process of LRs clustering. The present study was designed to
test this hypothesis to demonstrate that lysosomal ASMase
contributes to the formation of LR redox signaling platforms.
More specifically, when endothelial cells (ECs) are stimulated
by death-receptor agonists such as FasL or CD95 antibody and
TNF-, lysosomal vesicles will traffic and migrate into the
cell membrane, enriching ASMase locally. These SMases hydrolyze SM to produce ceramide and lead to the formation of
ceramide-enriched macrodomains or LR platforms, which
cluster or recruit NAD(P)H oxidase subunits, increase O2
production, and ultimately result in endothelial dysfunction.
To our knowledge, our results provide the first direct evidence
supporting the view that lysosome-like vesicles are importantly involved in the formation of endothelial LR redox signaling platforms.
MATERIALS AND METHODS
Cell culture
Fresh bovine hearts were obtained from a local abattoir and
immediately transported to our laboratory. The epicardial circumflex and anterior descending coronary arteries were quickly
dissected and placed in RPMI 1640 supplemented with 5% fetal calf serum (FCS), 2% antibiotic–antimycotic solution, 0.3%
gentamycin, and 0.3% nystatin and cleaned of surrounding adherent fat and connective tissue. The lumen of arterial segments
was filled with 0.25% collagenase A in RPMI 1640 supplemented with 0.1% bovine serum albumin (BSA) and incubated
at 37°C for 15–30 min. The arteries were then flushed with
RPMI 1640 supplemented with 2% antibiotic–antimycotic solution, 0.3% gentamycin, and 0.3% nystatin. Detached bovine
CAECs were collected and maintained in RPMI 1640 supplemented with 20% FCS, 1% glutamine, and 1% antibiotic–antimycotic solution at 37°C in 5% CO2. Isolated bovine CAECs
were identified by morphologic appearance (i.e., cobblestone
array) and by positive staining for von Willebrand factor antigen. All biochemical studies were performed by using CAECs
of two to four passages.
Confocal analysis of LR clusters in CAECs
Individual LRs are too small (50 nm) to be resolved by
standard light microscopy; however, if raft components are
crosslinked in living cells, clustered raft protein and lipid components can be visualized by fluorescence microscopy. For microscopic detection of LR platforms, CAECs were grown on
glass coverslips and then treated with 10 ng/ml FasL (UpstateMillipore, Billerica, MA) for 15 min to induce clustering of
lipid rafts. In additional groups of cells, bafilomycin A1 (Sigma,
St. Louis, MO; Baf, 100 nM) and glycyl-L-phenylalanine-naphthylamide (Sigma; MO, GPN, 100 M) were added to pretreat the cells for 15 min before FasL stimulation. These cells
were then washed in cold PBS and fixed for 10 min in 4% paraformaldehyde (PFA) in PBS and blocked with 1% BSA in TBS
for 30 min. GM1 gangliosides enriched in LRs were stained by
Alexa488-labeled cholera toxin B (alexa488-CTX, 1 g/ml, 45
min) (Molecular Probes, Eugene, OR). Cells were extensively
washed in cold PBS, fixed in 4% PFA for another 10 min, and
mounted on glass slide with Vectashield mounting media (Vector Laboratories, Inc., Burlingame, CA). Staining was visualized by using an Olympus scanning confocal microscope
(Olympus, Tokyo, Japan) at excitation/emission of 495/519 nm.
The patch or macrodomain formation of Alexa488-labeled CTX
and gangliosides complex represents the clusters of LRs. Clustering was defined as one or several intense spots or patches,
rather than the diffusion of fluorescence on the cell surface,
whereas a vast majority of unstimulated cells displayed a homogeneous distribution of fluorescence throughout the membrane. In each experiment, the presence or absence of clustering in samples of 200 cells was scored by unwitting researchers
independently after specifying the criteria for positive spots of
fluorescence. Cells displaying a homogeneous distribution of
fluorescence were marked negative. Results were given as the
percentage of cells showing one or more clusters after the indicated treatment as described.
Colocalization of LR clusters and NAD(P)H
oxidase subunits or ASMase in CAECs
For dual-staining detection of the colocalization of LRs and
NAD(P)H oxidase subunits, gp91phox, p47phox, or ASMase, the
CAECs were first incubated with Alexa488-labeled CTX as described earlier and then with mouse anti-gp91phox monoclonal
antibody (BD Biosciences, San Jose, CA; 1:200), mouse antip47phox monoclonal antibody (BD Biosciences; 1:200), or rabbit anti-ASMase polyclonal antibodies (Santa Cruz, Santa Cruz,
CA; 1:200), separately, which was followed by Texas red–conjugated anti-mouse or anti-rabbit (Molecular Probes, Eugene,
OR) secondary antibody as needed, respectively. An excitation/emission wavelength of 570/625 nm was used for confocal microscopy of Texas red.
Detection of lysosome trafficking to LR clusters
in CAECs
For these experiments, CAECs were plated in a 35-mm dish
(Nunclon, Raskilde, Denmark) at a density of 1 105 cells/ml
and then cultured for 1 day. Living CAECs were loaded with
150 nM LysoTracker Red DND-99 (Molecular Probes) and 2
1419
ENDOTHELIAL LIPID RAFT REDOX SIGNALING PLATFORMS
g/ml Alexa488-conjugated CTX for 30 min. In additional
groups of experiments, FasL, BAF, or GPN was added to the
final concentrations indicated in individual experiments. After
washing with PBS 3 times, the cells were visualized under a
confocal microscope with an excitation wavelength of 577 nm
and emission wavelength of 590 nm. In another group of cells,
LysoTracker colocalization with CTX was detected after incubation of both staining markers in these living cells.
Measurements of reactive oxygen species (ROS)
production in living CAECs
Detection of intracellular ROS was performed by a previously established method with a ROS-sensitive fluorescent
probe, 2,7-dihydrodichlorofluorescin diacetate (DCF) and
confocal microscopy (35). As described earlier, CAECs in 35mm dishes were washed with PBS and loaded with 20 M DCF
(Molecular Probes) for 30 min at 37°C. The intensity of fluorescence of ROS-reactive dichlorofluorescin was quantified by
using a laser-scanning confocal microscope (Olympus, Tokyo,
Japan) with excitation and emission wavelengths of 488 and
520 nm, respectively.
ESR detection of O2
ESR detection of O2 was also performed, as we described
previously (35). In brief, gently collected CAECs were suspended in modified Krebs/HEPEs buffer containing deferroximine (25 M, metal chelator). Approximately 1 106 CAECs
were mixed with 1 mM spin-trap 1-hydroxy-3-methoxycarbonyl-2,2,-5,5-tetrame-thyl-pyrrolidine (CMH) in the presence
or absence of 100 units/ml polyethylene glycol (PEG)-conjugated superoxide dismutase (SOD). The cell mixture loaded in
glass capillaries was immediately analyzed by ESR (Noxygen
Science Transfer & Diagnostics GmbH, Denzlingen, Germany)
for production of O2 at each minute for 10 min. The ESR settings were as follows: biofield, 3,350; field sweep, 60 G; microwave frequency, 9.78 GHz; microwave power, 20 mW;
modulation amplitude, 3 G; 4,096 points of resolution; receiver
gain, 500; and kinetic time, 10 min. The SOD-inhibitable signals were normalized by protein concentration and compared
among different experimental groups.
was obtained, cumulative dose–response curves to the endothelium-dependent vasodilator bradykinin (BK, 1010–106 M)
were determined by measuring changes in internal diameter. In
other groups of experiments, Baf or GPN was added into the
lumen of the artery to preincubate for 15 min before the subsequent application of FasL. The vasodilator response was expressed as the percentage relaxation of U-46619–induced precontraction based on changes in arterial internal diameter.
Internal arterial diameter was measured with a video system
composed of a stereomicroscope (Leica MZ8), a charge-coupled device camera (KP-MI AU, Hitachi), a video monitor
(VM-1220U, Hitachi), a video measuring apparatus (VIA-170;
Boeckeler Instrument), and a video printer (UP890 MD, Sony).
Statistics
Data are presented as mean SEM. Significant differences between and within multiple groups were examined by using ANOVA
for repeated measures, followed by Duncan’s multiple-range test.
A value of p 0.05 was considered statistically significant.
RESULTS
Detection of LRs clustering and aggregated
or recruited signaling molecules in the
membrane of CAECs
As shown in Fig. 1, left panel, typical LR patches in a CAEC
under resting condition and during FasL stimulation were de-
Isolated perfused small coronary
artery preparation
Fresh bovine hearts were obtained from a local abattoir.
Small coronary arteries (200 m ID) were prepared as we described previously (34). After dissection, arteries were transferred to a water-jacketed perfusion chamber and cannulated
with two glass micropipettes at their in situ length. The outflow
cannula was clamped, and the arteries were then pressurized to
60 mm Hg and equilibrated in physiologic saline solution (PSS,
pH 7.4) containing (in mM): NaCl, 119; KCl, 4.7; CaCl2, 1.6;
MgSO4, 1.17; NaH2PO4, 1.18; NaHCO3, 2.24; EDTA, 0.026;
and glucose, 5.5 at 37°C. PSS in the bath was continuously bubbled with a gas mixture of 95% O2 and 5% CO2 throughout the
experiment. After a 1-h equilibration period, the arteries were
precontracted by 50% of their resting diameter with a thromboxane A2 analogue, U-46619. Once steady-state contraction
FIG. 1. Representative confocal microscopy of LR clusters
(Alexa488-CTX green fluorescence at the left) and gp91phox
labeling (Texas red-conjugated anti-gp91phox at the middle). The overlay of two images exhibits yellow spots or
patches (right), which represent LRs clustering or colocalization of gp91phox and the LR component, ganglioside GM1.2.
1420
JIN ET AL.
tected by confocal microscopy. Under resting condition (control), only a diffuse fluorescent staining was observed on the
cell membrane, indicating possible distribution of single LRs.
However, when these cells were incubated with FasL, some
large fluorescent dots were shown on the cell membrane, indicating LR patches or macrodomains. In the presence of either
Baf or GPN, however, FasL-induced formation of LR clusters
was substantially attenuated, and in many cells, no LR patches
were detectable.
In the middle panel of Fig. 1, the cell stimulated by FasL
was scanned with an excitation/emission of 570/625 nm for
Texas red, which was conjugated to secondary antibody
against anti-gp91phox primary antibody. Red fluorescence indicated gp91phox positive. In addition to some diffuse signals
under resting conditions, relatively intense staining was found
on the cell membrane of the cell stimulated by FasL. When
two sequentially scanned images from the same cell with different wavelengths were merged, a number of yellow areas
were noted, either as dots or as patches resulting from green
CTX and Texas red-antiboby (right panel). These yellow
patches were considered the colocalization of LR clusters and
related molecules recognized by specific antibody. In this
case, NAD(P)H oxidase subunit gp91phox was the molecule
colocalized with LR clusters. When the cells were pretreated
by either Baf or GPN, both CTX clusters and aggregated Texas
red fluorescence were no longer observed, and no co-staining
of CTX and gp91phox was seen.
A
Positive cells of colocalization (%)
*
60
40
#
20
#
Ctrl Vehicle Baf GPN
FasL
D
70
60
p47phox
*
50
40
30
20
#
#
10
0
Ctrl Vehicle Baf GPN
FasL
70
gp91phox
*
60
50
40
30
20
#
#
10
0
Ctrl Vehicle Baf GPN
FasL
Positive cells with colocalization (%)
Positive cells with CTX (%)
80
CTX
0
Positive cells with colocalization (%)
For quantitative detection of LR clusters, CAECs were
stained only with Alexa488-CTX, which could allow a more
precise count of the large LR patches compared with doublestaining. The results for these experiments are summarized in
Fig. 2A. It was found that even control CAECs displayed a
small percentage of cells with LRs clustering (9.3 1.3%). After CAECs were stimulated with FasL, LRs-clustering positive
cells increased significantly to 73.3 8.4%; p 0.05; n 6).
When Baf or GPN was used to pretreat cells, the FasL-induced
increase in LRs-clustering positive cells was significantly reduced to 13.8 1.7%, or 16.6 3.1%, respectively. Figure 2B
summarizes the results of doubled-stained CAECs by
Alexa488-CTX and Texas red–conjugated gp91phox antibody.
These cells with yellow dots or patches after merging of fluorescent images with Alexa488 and Texas red indicated the colocalization of LR clusters and gp91phox. In control CAECs, only
about 5.5 3.3% cells displayed colocalization of LR clusters
and gp91phox. After the cells were treated with FasL, the percentage of colocalization-positive cells significantly increased
to 54.8 9.2%. Consistent with the results from those cells
stained by only CTX, pretreatment of CAECs with Baf or GPN
significantly reduced the colocalization-positive cells to 9.5 3.2% and 10.5 4.3%, respectively. Figure 2C shows the summarized results of co-staining of CAECs with CTXB and anti-
B
100
C
Quantified LR clusters and its colocalization with
NAD(P)H oxidase subunits or ASMase
100
ASM
*
80
60
40
20
#
#
0
Ctrl Vehicle Baf GPN
FasL
FIG. 2. Quantitative detection of
LR clusters and colocalization of
NAD(P)H oxidase subunits or ASMase in CAECs stimulated by
FasL. (A) Percentage changes in
positive cells stained by CTX during
FasL stimulation with or without
pretreatment by Baf or GPN. (B)
Percentage changes in positive cells
co-stained by CTX and anti-gp91phox
antibody during FasL stimulation
with or without pretreatment by Baf
or GPN. (C) Percentage changes in
positive cells co-stained by CTX and
anti-p47phox antibody during FasL
stimulation with or without pretreatment by Baf or GPN. (D) Percentage
changes in positive cells co-stained
by CTX and anti-ASMase antibody
during FasL stimulation with or
without pretreatment by Baf or GPN.
The values in all panels are expressed as mean SEM over 1,000
cells. (*p 0.05 vs. control; #p 0.05 vs. FasL group; n 6 primary
cultures of CAECs.)
ENDOTHELIAL LIPID RAFT REDOX SIGNALING PLATFORMS
1421
P47phox. Compared with control, the percentage of colocalization-positive cells in FasL-stimulated group increased significantly (51.8 9% vs. 6.5 3.1%; p 0.05; n 6). When the
cells were pretreated with Baf or GPN, the percentage of colocalization-positive cells, even in the presence of FasL, decreased to 8.5 3.5% and 8.8 3.3%, respectively. Figure 2D
shows the percentage of colocalization-positive cells in doublestaining experiments with CTX and anti-ASMase. Similarly,
FasL induced a significantly increase in colocalization-positive
cells, whereas after the cells were pretreated with Baf or GPN,
Fas L no longer significantly increased colocalization-positive
cells.
Association of LR clusters and lysosomal vesicles
in living cells
By using LysoTracker as a lysosome marker, we determined
the association of LRs clustering with lysosomal vesicles. Given
that LysoTracker is designed to detect pH-related function of
lysosomal vesicles in living cells, we first used a confocal microscope to examine whether time-dependent movements of
lysosomal vesicles occur toward the cell membrane. However,
we failed to observe obvious physical movements of lysosomal
vesicles in these CAECs stimulated by FasL, even with a frequent acquisition of fluorescent images every 2 s over a 15-min
period, including before and after the cells were treated with
FasL (Fig. 3).
However, when a double staining of these CAECs with LysoTracker and Alexa488-CTX was examined, images of the cells
exposed to FasL that were obtained by sequential scanning with
LysoTracker wavelengths and Alexa488 wavelengths showed
that FasL did not induce detectable increases or decrease in
lysosomes (Fig. 4, middle panel). However, CTX staining significantly increased when the cells were incubated with FasL
(Fig. 4, left panel). By merging two images, yellow areas were
detected in response to FasL stimulation, suggesting possible
colocalization of lysosomal vesicles with LR clusters (Fig. 4,
FIG. 4. Association of LR clusters and lysosomal vesicles in
live CAECs. Representative images were obtained by confocal microscopy with LR marker Alexa488-CTX on the left
(green) and colocalization of lysosomal vesicles labeled by
LysoTracker (red). The overlay of two images exhibits yellow
patches (right), which represent colocalization of lysosomal
vesicles and LRs. This was consistently observed in six batches
of primary cultures of CAECs.
right panel). In CAECs pretreated with Baf or GPN, however,
lysosome staining by LysoTracker was substantially attenuated,
accompanied by blockade of LRs clustering on the cell membrane, as shown in those images with labels of BafFasL and
GPNFasL. The quantitative percentage of LR cluster–positive cells is shown in Fig. 5. Similar to that observed in fixed
cells, after stimulation by FasL, the percentage of LR cluster–positive cells significantly increased (80.8 2.8% vs.
11.5 3.1%; p 0.05; n 6); when Baf and GPN were used
prior to FasL treatment, the percentage significantly decreased
to 14.0 1.8% and 13.5 1.9%, respectively.
Effects of Baf and GPN on FasL-induced
O2 production
FIG. 3. Detection of lysosome movement before and after
FasL stimulation. After loading with LysoTracker, the
CAECs were underwent a highly frequent acquisition of fluorescent images every 2 s over a 15-min period, including before and after the cells were treated with FasL. The location of
lysosomes in the cell before and 15 min after FasL stimulation
is shown. This was repeatedly observed in four separate experiments.
The direct consequence of the activation of NAD(P)H oxidase is increased production of O2. In Fig. 6A, typical ESR
spectra were shown as SOD-inhibitable O2 signals, and the
amplitude in each spectrum reflected the signal intensity. FasL
increased the amplitude of this SOD-inhibitable signal, as
shown by increases in amplitude. In the presence of Baf or GPN,
FasL no longer increased the amplitude of ESR spectra. Figure
6B presents the summarized data showing the fold changes of
O2 production. After these CAECs were treated with FasL,
the O2 level detected increased significantly. When the cells
were pretreated with Baf or GPN, this FasL-induced increase
in O2 production was significantly reduced.
1422
JIN ET AL.
group of arteries treated with FasL. By transferring the dilation
percentage into inhibition degree to the maximal dilation, FasL
was found significantly to attenuate the coronary vasodilator
response to BK, with an inhibition degree of 52.6% at the maximal dose of BK used in this study. When these arteries were
pretreated with Baf or GPN, the inhibition action of FasL on
BK-induced vasodilation was reduced to 2.1% and 11.4%, respectively.
LR Cluster Percentile
100
80
*
60
40
20
#
#
DISCUSSION
0
Ctrl Vehicle Baf GPN
FasL
FIG. 5. Quantitative detection of LR clusters stimulated by
FasL in live CAECs. This summarizes the percentage changes
in live cells stained by CTX during FasL stimulation with or without pretreatment with Baf or GPN. (*p 0.05 vs. control; #p 0.05 vs. FasL group; n 6 primary cultures of CAECs.)
In the present study, we examined the role of lysosomal vesicles in mediating FasL-induced formation of LR-redox signal-
A
Representative superoxide spectrum
Control
Effects of Baf and GPN on FasL-stimulated ROS
production in single CAECs
Reversal of FasL-induced impairment of
endothelium-dependent vasodilation by
Baf or GPN
Endothelium-dependent vasodilation induced by BK was determined in isolated perfused small bovine coronary arteries before and after FasL treatment. Figure 8 shows that BK produced
a concentration-dependent increase in internal diameter of these
small coronary arteries. Incubation of the arteries with FasL (10
ng/ml perfused into the lumen) had no significant effect on the
basal arterial diameters, but markedly attenuated the BK-induced increase of arterial diameters. The inhibitory effect of
FasL was reversed by a 15-min preincubation of the arteries
with Baf or GPN. However, treatment of the arteries with Baf
or GPN alone had no significant effect on basal arterial diameters or BK-induced vasodilation at the same doses used in the
FasL
Baf+FasL
GPN+FasL
0
0
B
500 1000 1500 2000 2500 3000 3500 4000
Time(s)
2.0
*
1.8
superoxide production
(folds normalized to control)
Because the ESR measurements described earlier could detect O2 only from bulk-prepared cell lysates, it could not detect the production of O2 in single cells. Further experiments
were done to measure the production of ROS through lysosomeassociated mechanism in single living cells. DCF was loaded
into the cells, and then a rapid response of ROS production to
FasL was observed. Figure 7A displays the dynamic changes
in ROS production that was trapped by DCF. The intensity of
DCF was quantitated to reflect the level of ROS within CAECs.
In each group of cells, the fluorescence at a time point of 0, 1,
3, and 5 min was quantitated from acquired images. As summarized in Fig. 7B, in nonstimulated CAECs, the fluorescence
intensity increased slightly over the experimental period. After
FasL was added, however, the fluorescence intensity within individual cells significantly increased in a time-dependent manner. When the cells were pretreated with Baf or GPN, the FasLinduced increase in the fluorescence intensity was significantly
attenuated, which was at a level similar to that obtained from
those cells without FasL stimulation.
1.6
#
1.4
#
1.2
1.0
0.8
Ctrl
Vehicle
Baf
FasL
GPN
FIG. 6. Effects of Baf or GPN on FasL-induced O2 production within CAECs. (A) Representative ESR spectra
showing SOD-inhibitable O2 signals. (B) Summarized data
depicting changes in O2 production in CAECs with different
treatments (*p 0.05 vs. control; #p 0.05 vs. FasL group;
n 5).
1423
ENDOTHELIAL LIPID RAFT REDOX SIGNALING PLATFORMS
A
0 min
1 min
3 min
5 min
control
FasL
Baf+FasL
GPN+FasL
B
2.6
ROS production
(Folds normalized to control)
2.4
2.2
2.0
control
FasL
Baf+FasL
GPN+FasL
*
*
1.8
*
1.6
1.4
1.2
1.0
of NAD(P)H oxidase activity and endothelial function in coronary arteries.
The major goal of the present study was to determine whether
and how lysosomal vesicles are involved in the formation of
LR redox signaling platforms and thereby regulate endothelial
function. Lysosomes are membrane-bound organelles that originate from the Golgi apparatus and exist in the cytoplasm of all
eucaryotic cells. These cytoplasmic organelles contain several
dozens of acid hydrolases that are primarily responsible for intracellular digestion (5). With their intracellular digestion function, lysosomes can operate enzymatic digestion of endocytosed
materials (cell defense) and aged organelles (cell autophagy)
(7). They also play important roles in receptor-mediated endocytosis and mediate events of receptor recycling (20). More recently, lysosomal vesicles have been demonstrated to be responsible for exocytosis in nonsecretory cells (18), where these
vesicles can fuse with the plasma membrane to excrete the contents of the vesicle and to incorporate the vesicle membrane
components into the cell membrane (18). This lysosomal vesicle fusion to the cell membrane may be importantly involved
in ASMase translocation in the process of LRs clustering, because this enzyme is present primarily in the lysosomal membrane, and on stimulation, it can be detected in the outer leaflet
of the cell membrane (11). To determine directly whether lysosomes or their function contribute to LRs clustering and consequent formation of membrane signaling platforms, we first
examined the effects of lysosome-function inhibition on death
factor–induced formation of LR redox signaling platforms associated with the aggregation of NAD(P)H oxidase subunits.
FasL was used in these experiments because previous studies
demonstrated that it is one of the most potent death factors to
induce LRs clustering and activation of NAD(P)H oxidase (6,
32). With a commonly used nontoxic cholera toxin subunit B
(CTX-B) as a marker of LRs (2), FasL was found to stimulate
0.8
1min
3min
Time
5min
FIG. 7. Effects of Baf and GPN on FasL-induced ROS production within single CAECs. (A) Representative DCF fluorescent images showing dynamic changes in ROS levels within
CAECs. (B) Summarized data depicting dynamic changes in
ROS production in CAECs at an early time of FasL stimulation when the cells were treated with Baf or GPN. (*p 0.05
vs. control; #p 0.05 vs. FasL group; n 6.)
ing platforms in the coronary artery EC membranes and corresponding changes in endothelial function in these arteries. By
using two structurally different lysosomal vesicle inhibitors, Baf
and GPN, we successfully blocked FasL-induced formation of
LRs clustering and aggregation of NAD(P)H oxidase subunits
in these LR clusters of CAECs, which resulted in significant
reduction of FasL-induced ROS production. In isolated perfused
small coronary arteries, FasL-induced impairment of vasodilator response to BK was also abolished by both vesicle inhibitors. These results strongly support the view that lysosomal vehicles contribute to the formation of LR redox signaling
platforms in CAECs and thereby participate in the regulation
100
control
FasL
Baf+FasL
GPN+FasL
80
60
Dilation (%)
basal
40
*
20
*
*
0
−11
−10
−9
−8
−7
BK (lg M)
−6
−5
FIG. 8. Effects of Baf and GPN on FasL-induced impairment of endothelium-dependent vasodilation to BK in isolated perfused small coronary arteries. *p 0.05 vs. control; n 5 bovine hearts.
1424
markedly the formation of LR clusters in bovine CAEC membrane (Figs. 1 and 2), which was consistent with previous results (32, 33). This FasL-stimulated LRs clustering was accompanied by aggregation of NAD(P)H oxidase subunits,
gp91phox, and recruitment of p47phox, constituting a redox signaling platform. Two mechanistically different lysosomal vesicle inhibitors, Baf and GPN, substantially blocked LRs clustering and accompanied aggregation or recruitment of
NAD(P)H oxidase subunits, suggesting that this redox molecule–associated LRs clustering is dependent on the integrity of
lysosomal structure or function.
It has been reported that Baf is a macrolide antibiotic that
specifically inhibits vacuolar H-ATPase (V-H-ATPase), resulting in a failure to pump protons into the lysosomal lumen
for acidification (3). Decreased acidification of the lysosomal
lumen has been shown to lead to loss of many lysosome functions, as listed earlier. Most important, Baf-damaged exocytosis and fusion of lysosomal membrane into the cell membrane
might block translocation of ASMase into the cell membrane
and thereby decrease ceramide production, plaguing LRs clustering associated with NAD(P)H oxidase. Indeed, we observed
a reduced ASMase colocalized with LR clusters in these ECs.
However, evidence exists that V-H-ATPase is present not only
in lysosomes or lysosome-related vesicles, but also in the
plasma membrane of certain cells such as endothelial cells (29).
It is a concern whether the effects of Baf observed in the present study are associated with the inhibition of this plasma membrane V-H-ATPase. Although recent studies have demonstrated that Baf-inhibitable V-H-ATPase activity is expressed
only in microvascular ECs, rather than in macrovascular ECs
(22), such as we used in the present study, we performed further experiments to address this concern.
Using another mechanistically different inhibitor of lysosome
function, GPN, the role of lysosomal vesicles in FasL-induced
LRs clustering associated with NAD(P)H oxidase subunits was
determined. GPN as a substrate of lysosomal cathepsin C may
be hydrolyzed, and its hydrolysis products could be accumulated in the vesicle, resulting in a reversible osmotic swelling,
and in this way interfering with lysosomal functions (1). As
shown in Figs. 1 and 2, GPN was shown to inhibit FasL-induced LRs clustering and colocalization of gp91phox and
p47phox, which was similar to that observed in Baf-treated cells.
Correspondingly, GPN also blocked an accumulation of ASMase in the LR clusters (Fig. 2D). These results further confirm that lysosomal vesicles do participate in the formation of
LR clusters with NAD(P)H oxidase subunits. The translocation
of lysosomal ASMase may be of importance in this lysosomerelated LRs clustering.
Further to explore the role of lysosomal vesicles in the formation of FasL-induced LR redox signaling platforms, we performed additional experiments to assess the movement of lysosomal vesicles to the cell periphery after stimulation with FasL
in living bovine CAECs by using confocal microscopy with
LysoTracker as a probe. LysoTracker is a fluorescent acidotropic probe for labeling and tracking lysosomal acidic organelles and has been widely used in studying trafficking of
lysosomal vesicles within various cells DIN (25, 30). However,
even with a very frequent acquisition of confocal fluorescent
images, we could not detect obvious trafficking of lysosomal
vesicles to the cell periphery in response to FasL stimulation.
JIN ET AL.
Given that LysoTracker stains primarily the intact lysosomal
vesicles with a luminal acidic environment, these direct observations of vesicle numbers by only using LysoTracker may not
reflect the dynamic changes in these vesicles under cell membrane because some vesicles trafficking to membrane might be
broken and fused and therefore could not be detected by single
LysoTracker staining.
To explore this possibility, we further performed live cell experiments with double staining of cells with LysoTracker and
Alexa488-CTX. It was found that FasL stimulation of CAECs
resulted in enhanced colocalization of LysoTracker-stained
lysosomal vesicles and Alexa488-CTX–labeled ganglioside
GM1.2, indicating that on stimulation, lysosomal vesicles are
closer to membrane components or LRs, where they may fuse
and incorporate ASMase into the cell membrane, promoting
ceramide production and formation of LR platforms. This
ASMase relocation and ceramide production in the cell plasma
membrane has been reported in a number of previous studies
(2). In addition, the colocalization of lysosomal vesicles and
LR components was significantly blocked by treatment with
Baf and GPN, further suggesting that functional lysosomal vesicles are present in the proximal sites to LRs.
These lysosomal vesicles may interact with the cell membrane when the cells are stimulated by certain factors such as
FasL. In previous studies, lysosomal displacement to the cell
periphery was indicated as a mechanism to facilitate increased
secretion of degradative enzymes such as matrix metalloproteinases, serine proteases, and cathepsins that are sequestered
in lysosomal vesicles. This trafficking of lysosomal vesicles
would be the first step in the recruitment of lysosomes toward
the cell surface before lysosomal exocytosis (10, 24). However,
this displacement of lysosomal vesicles is often observed in
some cells subject to brief or relatively long stimulation (25,
30). Because we attempted to examine the early mechanism by
which LR redox signaling platforms are formed in response to
activation of death receptors, a relatively short exposure of cells
to death factors such as FasL was used in our experiments.
Therefore, the role of lysosomal vesicles in LRs clustering observed under this experimental condition may be attributed
mainly to those vesicles in the proximity of cell membrane.
Nevertheless, a trafficking of lysosomal vesicles in these
CAECs during a brief or long-term treatment of FasL could not
be excluded.
Next, we examined whether lysosomal vesicles contribute to
the production of O2 from LR platforms with NAD(P)H oxidase subunits. CAECs were stimulated by FasL in the absence
and presence of Baf or GPN, and then intracellular O2 production was detected by ESR spectrometry. In the absence of
Baf or GPN treatment, FasL significantly increased O2 production in CAECs. However, this FasL-induced O2 production was markedly reduced when the cells were treated by either Baf or GPN. These results indicate that LRs clustering
during FasL stimulation is indeed accompanied by activation
of NAD(P)H oxidase. Although these experiments could not
make sure that O2 production in response to FasL is from
cell-membrane LR platforms, our previous studies demonstrated in the same experimental condition that activation of
NAD(P)H oxidase during FasL stimulation occurred mainly in
LR membrane fractions (32). In addition, we monitored a dynamic change in O2 level in single cells by fluorescent imag-
1425
ENDOTHELIAL LIPID RAFT REDOX SIGNALING PLATFORMS
ing analysis of ROS within the cells. It was shown that, even
at the first minute after FasL stimulation, ROS within CAECs
were significantly increased. Pretreatment of these cells with
either Baf or GPN completely blocked increases in intracellular O2 levels. It appears that lysosomal vesicles are involved
in the activation of NAD(P)H oxidase, which was consistent
with the time course of the LRs clustering formation in response
to FasL stimulation (10–20 s), as demonstrated in previous studies (13, 17). The results from a more recent study that endosomal acidification through V-H-ATPase during Fas activation
enhanced ceramide formation and activation of NAD(P)H oxidase also support the possible interactions of lysosome-related
vesicles and NAD(P)H oxidase activity (21). Taken together,
these findings provide evidence that lysosomal vesicles importantly contribute to the activation of NAD(P)H oxidase through
ceramide production and consequent formation of LR platforms.
The functional significance of this lysosomal vesicles–mediated formation of LR signaling platforms and activation of
NAD(P)H oxidase was tested by measurement of endotheliumdependent vasodilation in isolated perfused small coronary arteries. It has been reported that endothelium-dependent vasodilation represents an early functional change in response to
activation of death-receptor activation (31–33). In those studies, the role of LR clusters, ceramide, ASMase, and NAD(P)H
oxidase in mediating endothelial dysfunction induced by various death factors was demonstrated. The present study further
addressed whether lysosomal vesicles are also involved in the
impairment of endothelial function associated with death-receptor activation. Similar to our previous studies (32, 33), FasL
was found significantly to attenuate coronary vasodilator response to BK, a potent endothelium-dependent vasodilator in
coronary circulation. When these arteries were pretreated with
Baf or GPN, the blunting action of FasL on BK-induced vasodilation was almost completely blocked. This provides clear evidence that the functional integrity of lysosomal vesicles in the
coronary arterial endothelium is important to the action of FasL
on endothelial function. Based on the results discussed earlier
regarding the role of lysosomal vesicles in the formation of LR
redox signaling platforms and activation of NAD(P)H oxidase,
we believe that failure of FasL to cause early endothelial dysfunction in Baf- or GNP-treated arteries may be associated with
less production of O2 production from the LR clustered
NAD(P)H oxidase complex. Although the effect of NADPH
oxidase inhibitor on the FasL-induced dilatation response were
not directly tested, in a previous study (32) from our group, we
confirmed this action of NADPH oxidase inhibitors to the FasLinduced impairment of vasodilation, which together supports
the action of NADPH oxidase–derived ROS in this process. The
role of this reduced production of O2 in improvement of endothelial function during activation of death receptors has been
shown in several of our previous studies (31–33) and by others (4, 8).
In summary, the present study demonstrated that lysosomal
vesicles in the proximity of the cell membrane may play an important role in the formation of LR signaling platforms. If the
integrity of lysosomal function or structure is lost, as shown in
CAECs treated with Baf or GPN, LRs clustering and consequent aggregation or recruitment of NAD(P)H oxidase subunits
could be blocked, thereby suppressing the actions of activated
death receptors to transmit signals to downstream targets. This
lysosomal vesicles–mediated formation and activity of LR redox signaling platforms may represent an important early mechanism responsible for endothelial dysfunction induced by death
factors.
ACKNOWLEDGMENTS
This study was supported by grants from the National Institutes of Health (HL-57244, HL-75316, and DK54927). We
thank Dr. Ningjun Li for his kindly help in writing the manuscript.
ABBREVIATIONS
DCF, 2,7-dihydrodichlorofluorescin diacetate; ASMase,
acid sphingomyelinase; Baf, bafilomycin; BK, bradykinin;
BSA, bovine serum albumin; CTXB, cholera toxin subunit B;
CAECs, coronary artery endothelial cells; EGFR, epidermal
growth factor receptor; FasL, Fas ligand; GPN, glycyl-Lphenylalanine--naphthylamide; LR, lipid raft; PFA, paraformaldehyde; PBS, phosphate-buffered saline; ROS, reactive
oxygen species; SM, sphingomyelin; O2, superoxide; TNFR,
tumor necrosis factor receptor.
REFERENCES
1. Berg TO, Stromhaug E, Lovdal T, Seglen O, and Berg T. Use of
glycyl-L-phenylalanine 2-naphthylamide, a lysosome-disrupting
cathepsin C substrate, to distinguish between lysosomes and
prelysosomal endocytic vacuoles. Biochem J 300: 229–236, 1994.
2. Bollinger CR, Teichgraber V, and Gulbins E. Ceramide-enriched
membrane domains. Biochim Biophys Acta 1746: 284–294, 2005.
3. Bowman EJ, Siebers A, and Altendorf K. Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells,
and plant cells. Proc Natl Acad Sci U S A 85: 7972–7976, 1988.
4. Cardier JE and Erickson-Miller CL. Fas (CD95)- and tumor necrosis factor-mediated apoptosis in liver endothelial cells: role of
caspase-3 and the p38 MAPK. Microvasc Res 63: 10–18, 2002.
5. Cutler DF. Introduction: lysosome-related organelles. Semin Cell
Dev Biol 13: 261–262, 2002.
6. Eramo A, Sargiacomo M, Ricci-Vitiani L, Todaro M, Stassi G,
Messina CG, Parolini I, Lotti F, Sette G, Peschle C, and De Maria
R. CD95 death-inducing signaling complex formation and internalization occur in lipid rafts of type I and type II cells. Eur J Immunol 34: 1930–1940, 2004.
7. Eskelinen EL. Roles of LAMP-1 and LAMP-2 in lysosome biogenesis and autophagy. Mol Aspects Med 27: 495–502, 2006.
8. Filippatos G, Ang E, Gidea C, Dincer E, Wang R, and Uhal BD.
Fas induces apoptosis in human coronary artery endothelial cells
in vitro. BMC Cell Biol 5: 6, 2004.
9. Furne C, Corset V, Herincs Z, Cahuzac N, Hueber AO, and Mehlen
P. The dependence receptor DCC requires lipid raft localization for
cell death signaling. Proc Natl Acad Sci U S A 103: 4128–4133,
2006.
10. Glunde K, Guggino SE, Solaiyappan M, Pathak AP, Ichikawa Y,
and Bhujwalla ZM. Extracellular acidification alters lysosomal trafficking in human breast cancer cells. Neoplasia 5: 533–545, 2003.
11. Goggel R, Winoto-Morbach S, Vielhaber G, Imai Y, Lindner K,
Brade L, Brade H, Ehlers S, Slutsky AS, Schutze S, Gulbins E,
and Uhlig S. PAF-mediated pulmonary edema: a new role for acid
sphingomyelinase and ceramide. Nat Med 10: 155–160, 2004.
1426
12. Grassme H, Jekle A, Riehle A, Schwarz H, Berger J, Sandhoff K,
Kolesnick R, and Gulbins E. CD95 signaling via ceramide-rich
membrane rafts. J Biol Chem 276: 20589–20596, 2001.
13. Gulbins E and Kolesnick R. Acid sphingomyelinase-derived ceramide signaling in apoptosis. Subcell Biochem 36: 229–244, 2002.
14. Gupta N and DeFranco AL. Visualizing lipid raft dynamics and
early signaling events during antigen receptor-mediated B-lymphocyte activation. Mol Biol Cell 14: 432–444, 2003.
15. Holopainen JM, Subramanian M, and Kinnunen PK. Sphingomyelinase induces lipid microdomain formation in a fluid
phosphatidylcholine/sphingomyelin membrane. Biochemistry 37:
17562–17570, 1998.
16. Jury EC and Kabouridis PS. T-lymphocyte signalling in systemic
lupus erythematosus: a lipid raft perspective. Lupus 13: 413–422,
2004.
17. Kirschnek S, Paris F, Weller M, Grassme H, Ferlinz K, Riehle A,
Fuks Z, Kolesnick R, and Gulbins E. CD95-mediated apoptosis in
vivo involves acid sphingomyelinase. J Biol Chem 275:
27316–27323, 2000.
18. Moreno RD and Alvarado CP. The mammalian acrosome as a secretory lysosome: new and old evidence. Mol Reprod Dev 73:
1430–1434, 2006.
19. Ni X and Morales CR. The lysosomal trafficking of acid sphingomyelinase is mediated by sortilin and mannose 6-phosphate receptor. Traffic 7: 889–902, 2006.
20. Nurminen TA, Holopainen JM, Zhao H, and Kinnunen PK. Observation of topical catalysis by sphingomyelinase coupled to microspheres. J Am Chem Soc 124: 12129–12134, 2002.
21. Reinehr R, Becker S, Braun J, Eberle A, Grether-Beck S, and
Haussinger D. Endosomal acidification and activation of NADPH
oxidase isoforms are upstream events in hyperosmolarity-induced
hepatocyte apoptosis. J Biol Chem 281: 23150–23166, 2006.
22. Rojas JD, Sennoune SR, Maiti D, Bakunts K, Reuveni M, Sanka
SC, Martinez GM, Seftor EA, Meininger CJ, Wu G, Wesson DE,
Hendrix MJ, and Martinez-Zaguilan R. Vacuolar-type HATPases at the plasma membrane regulate pH and cell migration
in microvascular endothelial cells. Am J Physiol Heart Circ Physiol 291: H1147–H1157, 2006.
23. Rojas JD, Sennoune SR, Martinez GM, Bakunts K, Meininger CJ,
Wu G, Wesson DE, Seftor EA, Hendrix MJ, and Martinez-Zaguilan
R. Plasmalemmal vacuolar H-ATPase is decreased in microvascular endothelial cells from a diabetic model. J Cell Physiol 201:
190–200, 2004.
24. Rozhin J, Sameni M, Ziegler G, and Sloane BF. Pericellular pH
affects distribution and secretion of cathepsin B in malignant cells.
Cancer Res 54: 6517–6525, 1994.
25. Sarnataro D, Grimaldi C, Pisanti S, Gazzerro P, Laezza C, Zurzolo
C, and Bifulco M. Plasma membrane and lysosomal localization
of CB1 cannabinoid receptor are dependent on lipid rafts and regulated by anandamide in human breast cancer cells. FEBS Lett 579:
6343–6349, 2005.
26. Sato K, Shikano S, Xia G, Takao J, Chung JS, Cruz PD Jr, Xie
XS, and Ariizumi K. Selective expression of vacuolar H-ATPase
JIN ET AL.
27.
28.
29.
30.
31.
32.
33.
34.
35.
subunit d2 by particular subsets of dendritic cells among leukocytes. Mol Immunol 43: 1443–1453, 2006.
Schissel SL, Keesler GA, Schuchman EH, Williams KJ, and Tabas
I. The cellular trafficking and zinc dependence of secretory and
lysosomal sphingomyelinase, two products of the acid sphingomyelinase gene. J Biol Chem 273: 18250–18259, 1998.
Sprenger RR, Fontijn RD, van Marle J, Pannekoek H, and Horrevoets AJ. Spatial segregation of transport and signalling functions between human endothelial caveolae and lipid raft proteomes.
Biochem J 400: 401–410, 2006.
Sun-Wada GH, Wada Y, and Futai M. Lysosome and lysosomerelated organelles responsible for specialized functions in higher
organisms, with special emphasis on vacuolar-type proton ATPase.
Cell Struct Funct 28: 455–463, 2003.
Trajkovic K, Dhaunchak AS, Goncalves JT, Wenzel D, Schneider
A, Bunt G, Nave KA, and Simons M. Neuron to glia signaling triggers myelin membrane exocytosis from endosomal storage sites. J
Cell Biol 172: 937–948, 2006.
Zhang AY, Teggatz EG, Zou AP, Campbell WB, and Li PL. Endostatin uncouples NO and Ca2 response to bradykinin through
enhanced O2* production in the intact coronary endothelium. Am
J Physiol Heart Circ Physiol 288: H686–H694, 2005.
Zhang AY, Yi F, Zhang G, Gulbins E, and Li PL. Lipid raft clustering and redox signaling platform formation in coronary arterial
endothelial cells. Hypertension 47: 74–80, 2006.
Zhang AY, Yi F, Zhang G, Gulbins E, and Li PL. RNA interference of acidic sphingomyelinase gene restores endothelial function
through inhibition of NAD(P)H oxidase activity. FASEB J 19:
A521, 2005.
Zhang DX, Yi FX, Zou AP, and Li PL. Role of ceramide in TNFalpha-induced impairment of endothelium-dependent vasorelaxation in coronary arteries. Am J Physiol Heart Circ Physiol 283:
H1785–H1794, 2002.
Zhang G, Zhang F, Muh R, Yi F, Chalupsky K, Cai H, and Li PL.
Autocrine/paracrine pattern of superoxide production through
NAD(P)H oxidase in coronary arterial myocytes. Am J Physiol
Heart Circ Physiol 292: H483–H495, 2007.
Address reprint requests to:
Pin-Lan Li, M.D, Ph.D.
Department of Pharmacology and Toxicology
Medical College of Virginia
Virginia Commonwealth University
410 N 12th, Richmond, VA 23298
E-mail: [email protected]
Date of first submission to ARS Central, March 27, 2007, date
of final revised submission, March 27, 2007; date of acceptance, March 28, 2007.
This article has been cited by:
1. Pin-Lan Li , Erich Gulbins . 2007. Lipid Rafts and Redox Signaling. Antioxidants & Redox Signaling 9:9, 1411-1416. [Abstract]
[PDF] [PDF Plus]